The Authors Reply

    loading  Checking for direct PDF access through Ovid


We would like to thank Labgaa et al for their interest regarding our article1 and the opportunity to respond to their comments.
First, receiver operating characteristic curve analysis was used to examine the predictive accuracy of Δalbumin at postoperative day (POD) 1. The cutoff value of Δalbumin at POD 1 was 13.8% with area under the curve (AUC) 0.753 and Youden index 0.410. The result indicated that Δalbumin with nadir albumin within POD 2 performed better than Δalbumin at POD 1 in our study.
Second, the performance of absolute vs relative Δalbumin was compared using receiver operating characteristic curve analysis. The cutoff value of absolute Δalbumin was 5.95 g/L in our study with AUC 0.782 and Youden index 0.472. When relative Δalbumin was used, AUC was 0.810 and Youden index was 0.510. The result indicated that relative Δalbumin performed better than absolute Δalbumin in predicting complications after colorectal resection.1 This might be explained by the individual variation of baseline albumin level before surgery.
Finally, we appreciate that Labgaa et al2 verified our finding that serum albumin is a promising biomarker to predict postoperative complications. The current study was submitted and published ahead of the article by Labgaa et al.2 Also, we could not find “surprising similarities” between our study and NCT02356484 after reviewing the protocol registered at,3 which evaluated the predictive value of albumin, C-reactive protein, procalcitonin, and lactates in terms of surgical stress and postoperative complications. The 2 studies were different in design (retrospective vs prospective), patient population, inclusion and exclusion criteria, data collection (nadir level within POD 2 vs POD 1–4 albumin level), and cutoff value, except for the similar topic. The concept of our study came from our observation in clinical practice and was inspired by the studies of Sang et al4 and Spolverato et al.5 In our article, we pointed out the study’s limitation as a retrospective observational study1 and also cited the article by Hübner et al.6 Labgaa et al2 did perform a good prospective cohort study and extrapolated the role of Δalbumin in other types of abdominal surgery besides colorectal resection. However, their speculation regarding the “surprising similarities” between the protocol of NCT023564843 and our study is inappropriate.
    loading  Loading Related Articles